Kaposi's sarcoma (KS) is a complex disease which has puzzled dermatologists and cell biologists for over 100 years. Once a rare and self-limited disorder, it is now the most common tumor in the AIDS patient population, in whom its manifestations are severe. This proposal is based on a novel murine model in which SV40 T antigen (ag) transformed microvascular endothelial cells induce lesions which bear a striking resemblance to human KS. It is hypothesized that the unusual tumorigenicity of these cells represents virally induced changes in endothelial cell (EC) gene regulation which alter the relationship between these cells and normal tissues. Studies are proposed to delineate the contributions of both genetic and epigenetic factors to the transition from transformation to tumorigenesis. The three principle components of these lesions are the transformed EC, normal host tissue cells, and an inflammatory infiltrate. Many of the functions served by normal EC are know to be critical determinants of tumorigenesis, but little is know about their behavior as cell or origin in malignant transformation. Understanding the biology of EC derived tumors, particularly KS, promises to expand and refine our concept of malignancy, a concept which has evolved largely from an emphasis on genetic changes in cancer cells. KS may serve to remind us that the "soil: is an important as the "seed". The following specific aims and approaches are proposed: (1) Identify and characterize EC specific gene products which are determinants of tumorigenicity. Experiments are planned to delineate the contributions of (a) autocrine and paracrine growth factors, (b) extracellular matrix (ECM), and (c) cellular transdifferentiation and tumor suppressor genes, using biochemical purification, antisense inhibition, and gene cloning. (2) Examine the immune response to KS-like tumors in mice, emphasizing the role of EC-derived factors in modulation of immune effector function. In vitro assays will be used to determine susceptibility to lysis of tumor derived clones. (3) Design and test innovative therapies in the murine model, based on pathophysiologic concepts developed in the specific aims above.